Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/cdeust/ai-architect-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/cdeust/ai-architect-mcp/generate-prd)<a href="https://agentmods.dev/commands/cdeust/ai-architect-mcp/generate-prd"><img src="https://agentmods.dev/badge/commands/cdeust/ai-architect-mcp/generate-prd.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00017 | $0.00418 |
| Opus 5 | $0.00009 | $0.00209 |
| Sonnet 5 | $0.00003 | $0.00084 |
| Haiku 4.5 | $0.00002 | $0.00042 |
Grade A, and why
generate-prd scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Generate PRD
Step 1 — Load Stage 4 skill
Read the PRD generation skill:
Read skills/stage-4-prd/SKILL.md
This defines the full PRD generation workflow including clarification, generation via ai-prd-generator, HOR verification, and evaluator-optimizer retry loop.
Step 2 — Verify MCP tools
Call the MCP server to verify connectivity:
ai_architect_select_strategy(task_type="prd_generation", context="generate PRD")
Step 3 — Execute
Follow the Stage 4 SKILL.md instructions:
- Load few-shot examples from prior successful PRDs
- Generate clarification questions for open gaps
- Run clarification loop until user says "proceed"
- Build enhanced prompt with selected strategy
- Delegate to ai-prd-generator plugin for 9-file generation
- Run 64 HOR rules on generated PRD
- Run verification algorithms (CoV, Graph, NLI)
- If compound score < 0.85: inject failures and retry (max 3)
If $ARGUMENTS is provided, use it as the feature description. Otherwise, ask for requirements.
Output
Nine files in the current directory:
prd-overview.md— Executive summaryprd-requirements.md— Functional and non-functional requirementsprd-user-stories.md— User stories with acceptance criteriaprd-technical.md— Technical specificationprd-acceptance.md— Acceptance criteriaprd-roadmap.md— Implementation roadmapprd-jira.md— JIRA ticketsprd-tests.md— Test casesprd-verification.md— Verification report
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 55 lines · 17 tokens per session scan A 71a8c5038864
generate-prd is a command published in the GitHub repository cdeust/ai-architect-mcp (1 stars, last pushed 4mo ago), licensed MIT. It adds 17 tokens to every session and 418 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
check
Run the full local quality gate — TypeScript check, build, dependency audit (prod-only), skill registry validate. Fast pre-commit / pre-suggestdeploy sweep.
ir
Incident response workflow — triage, evidence collection, timeline, and IOC extraction.
pwned
Post-exploitation workflow after getting shell access — privesc, credential harvest, lateral movement.
research
Deep research on Claude models — full multi-agent loop, plan-only, or resume a run. Persists cited evidence to runs/ /.
sp.reverse-engineer
Reverse engineer a codebase into SDD-RI artifacts (spec, plan, tasks, intelligence).
sp.checklist
Generate a custom checklist for the current feature based on user requirements.